your concerns are about why change is needed, and about whether change is safe. thats fair, and shows the care for the ecosystem.
believe me, thousands of tools, frameworks and platforms have gone through exactly this: Java, Spring, Hibernate, Angular, React, Django, Rails, .NET... you name it. and all of them shipped major versions that broke things and scared their ecosystems. companies
kept saying "we can't migrate right now." something broke and someone fixed it and the platform came out stronger. it is the normal lifecycle of healthy software.
and i say this from what i see. my day job is migration. i live in the middle of this friction every day. i can see that "companies will be scared off by v5", but they may be also scared off by the perception that Eclipse is not moving forward.
on not knowing the exact list of changes for v5: you are right, and i will own that. i like the change set emerging from the community, not dictated by one person. and this gets in to the depth of OSS culture:
"bring a PR and we will discuss it" (sounds open but is actually hard for people without knowledge about many repos with hundreds of bundles and their relations and inter dependencies and the overarching architecture and the prevailing tradeoffs)
"this is what Eclipse looks like today, this is why it feels slow and heavy, this is why users drift away, this is what we need to fix that (call it something else than v5 if that word scares folks) this is what the tradeoffs are, and this is the way forward.
commits welcome."
a shift from gatekeeping to beckoning. a shift from mailing list discussion into a contributor pipeline.
and thank you for the devcall. i always had the invite, but missed every time. will try to join next one.
This Message Is From an External Sender
This message came from outside your organization.
on resources: i agree on the constraints, but gently disagree on the inference. a v5 commitment is not a demand for more volunteer hours from existing contributors, but a clear message to companies that investment here has a direction and a destination. setting
a clear vision and roadmap can help drive more contributors.
How would you get contributors interested in doing that work that wouldn't be possible without a "v5 commitment"? Also note that I think that most companies earning money from Eclipse (and the RCP
applications they are building) would be scared off by hearing about a v5. Not getting enough resources "despite" being a rewrite/big new thing is pretty much the main problem Initiative 31 is facing so I don't see why this would be different.
on per-change compatibility windows: here our difference is very tiny! you describe multiple changes with individual migration windows moving at different speeds. the issue with that is, in the absence of overarching commitment to remove those, the layers can
become permanent, and Eclipse 4 is the evidence. the e4 compatibility layer had no coordinated sunset, so it is still there.
Many products would probably never migrate, this isn't really an option in my opinion.
on AI: my AI argument is about Eclipse's survival and relevance, not about forcing personal preferences or changing any individual's workflow.
I still have no idea why a rewrite would be necessary or even help with integrating AI when everything necessary is there. And AI shouldn't be the reason for a rewrite breaking compatibility anywhere
else. Anything done "for AI" shouldn't negatively impact anyone who doesn't want it. (I'm also tired of the "AI is necessary to survive" argument, you don't need to repeat it to me.)
the changes need to happen
If you are talking about agreeing on specific changes that need to happen, I don't see that. While there are some things I'd like to see (including work on CDS which doesn't require anything impacting
compatibility for example), I don't think there are that many changes relevant here that I think need to happen. As you mentioned wanting to set a "clear vision and roadmap", I don't think you even know the changes that "need to happen" which I'd argue is
a prerequisite to agreeing on these changes with anyone.
changes need clear migration paths
migration need explicit expiry
I'm not so sure about this. While I don't have an issue with migration paths, there are many reasons why plans like that need to change. If people didn't finish getting rid of a deprecated feature,
that might be an argument for keeping it (at least for some time) and people can still get the advantages of the new feature while the old version is still around. While it can be helpful to define a plan for when something should be removed, that would rather
be a "minimum" time.
rcp as a class I citizen.
I guess that is something I personally agree with (despite not working on any RCP applications).
To be honest, I also don't like the idea of a compatibility "layer", I prefer having building on top of existing things (e.g. capability to use JPMS modules within OSGi and e.g. generating module descriptors
at compile-time (tycho) and using the existing approach where necessary) without really deprecating the "old" approach.
As I previously mentioned with regard to scaring people with a v5, I would strongly recommend not calling it v5. I don't think people want that.
That being said, please note that I'm not a committer on any Eclipse project, I am not involved in decisions. I am just saying my personal opinions (which may or may not help you with improving your
proposal). However, in addition to the mailing lists, I think you might be interested in the devcall every two weeks:
https://github.com/eclipse-platform/eclipse.platform.releng.aggregator/wiki/Eclipse-IDE-%E2%80%90-Developers-Community-Call. There, some people working on Eclipse including some people working on RCP apps meet and you might be able to gather other opinions
on that topic. You can find an invite in the calendar linked on the top of that page. However, I would still recommend providing concrete evidence on why you think it's necessary to break compatibility in a significant way (at least I'm not convinced of that
at all).
On 29/05/2026 17:37, GIREESH PUNATHIL wrote:
Daniel: great! thankful to see you acknowledging i) competitive gap exists ii) lazy activation could eventually become default and iii) need of compatibility window. i think we are converging on many aspects!
on resources: i agree on the constraints, but gently disagree on the inference. a v5 commitment is not a demand for more volunteer hours from existing contributors, but a clear message to companies that investment here has a direction and a destination. setting
a clear vision and roadmap can help drive more contributors.
on per-change compatibility windows: here our difference is very tiny! you describe multiple changes with individual migration windows moving at different speeds. the issue with that is, in the absence of overarching commitment to remove those, the layers can
become permanent, and Eclipse 4 is the evidence. the e4 compatibility layer had no coordinated sunset, so it is still there.
on RCP: agree 100%. any v5 process should consider RCP as the exemplary use case when defining migration paths, deprecations, sunset and timelines. if a v5 migration path works for RCP, it works for everyone.
on AI: my AI argument is about Eclipse's survival and relevance, not about forcing personal preferences or changing any individual's workflow.
in summary, i think we agree on most parts:
- the changes need to happen
- changes need clear migration paths
- migration need explicit expiry
- rcp as a class I citizen.
where we may differ is whether the compatibility layer needs a coordinated migration path and expiry. to me, that definition is what a v5 draft could contain.
WDYT?
This Message Is From an External Sender
This message came from outside your organization.
> while each of these is technically valid, they all are proposals to optimize within the current architecture, not helping evolution based on changing technologies, programming practices and expectations.
That is the point I'm making. There are a lot of opportunities for improvement within the current architecture (and some with possibly major non-breaking changes). With people not having enough time/resources
to improve performance within the "current architecture", I don't have any reason to think why people would be willing to take on the even bigger effort of making major changes to how everything works. For example, you could take a look at Initiative 31 (https://github.com/swt-initiative31/documents/blob/main/report/overall_report.md)
which was about changing SWT to use a single (more modern) stack across operating systems. It concluded that it would be possible (some things are even being integrated - without significantly breaking compatibility) but it would be too much work without companies
willing to invest the necessary amount of resources.
> lazy activation? add a system property
That system property would be a compatibility layer that could eventually default to lazy activation (but using eager activation at first).
> here is my take: the traditional IDE shell (the menus, the views, perspectives, the edit-compile-debug loop) is under genuine pressure from AI tools. they are already replacing significant parts
of the classical developer workflow and making many of these views and actions redundant.
While this may be applicable to many people, there are also a lot of people that don't want to use AI tools. Any change made should not negatively affect that workflow (in my opinion).
Another thing you should consider is RCP applications which is one thing where Eclipse is used a lot (and which contributes a lot to the resources spent on developing the Eclipse IDE). With there being
many applications based on the Eclipse Platform, changes like that (especially something like replacing the class loading approach with JPMS modules) have a noticable (not to say disproportionate) affect on these applications as well whereas forcing them to
rewrite their application to improve the standard IDE is not reasonable.
To sum it up, I think it's probably better to make smaller changes in a compatible way (and deprecate and remove these compatibility helpers/layers in the future) rather than creating a v5 where there
are a lot of changes at once (which also won't be completed within the next few years). That would also allow moving faster with some parts of that effort than with others. Instead of one compatibility window, there can be a compatibility window for each change
requiring it.
On 28/05/2026 13:49, GIREESH PUNATHIL via eclipse-dev wrote:
at this point i want to step back from individual topics, as i think otherwise the top level point gets lost. every time a structural concern is raised, the response has been a local fix:
startup too slow? some JDK flag gives us 20%
lazy activation? add a system property
lingering stale APIs? policy is in place
while each of these is technically valid, they all are proposals to optimize within the current architecture, not helping evolution based on changing technologies, programming practices and expectations.
the competitive reality is that eclipse is losing ground[1]. it is not an option or an opinion, instead reality. AI SDK integrations ignored us. IMO, other IDEs didn't become popular through better Java support but with better architectural choices from day
one. The 20% startup improvement from AOT is real and great, but that 20% of 5 seconds is still 4. VSCode starts in 2. that gap is architectural and is not tunable.
my v5 proposal is not "rewrite everything and break the world.”. it is a request for a draft:
- what the next major version should look like
- a formal statement about architectural directions. with reasons and contexts
- a defined compatibility window for the old path [ example: apple migrated from ppc (p) to inel (x) to arm (a). easily. each time, a compatibility layer ran the old arch while the new one matured. and each time, there was a hard deadline. and each time the
ecosystem moved because it had to! ]
Stephan asked the most relevant and honest question: "does anybody still need us?" lets all try to answer that directly rather than deflect it. here is my take: the traditional IDE shell (the menus, the views, perspectives, the edit-compile-debug loop) is under
genuine pressure from AI tools. they are already replacing significant parts of the classical developer workflow and making many of these views and actions redundant.
This Message Is From an Untrusted Sender
You have not previously corresponded with this sender.
From what I recall, the "expected" startup improvement with these things for most applications wouldn't be much more than 20% so I'm not sure how much that would actually improve in that area. Also,
my point was that there may be a possibility to get some of that improvement by supporting externally loaded classes in some way.
If you think about improving startup time using lazy activation, you could try asking Heiko Klare (who participated in that aforementioned discussion) how changing the activation policy to lazy (or
eager) would affect their startup time since they are interested in better startup time. If you really want lazy bundle activation by default, I think the way to go would be creating a system property or similar that affects the default (and if some plugin
explicitly specifies the bundle activation policy, I guess that should still be respected).
However, I don't see any reason why breaking changes in a "v5" would motivate people to do performance work.
On 28/05/2026 06:58, GIREESH PUNATHIL wrote:
thank you Daniel for the link - impressive work and impressive numbers! but pls note that the 20% startup improvement with JDK assistance will further benefit if assisted by eclipse' own re-architecture: probably in order of multiples. for example VS code has
extension lazy loading by default, enforced at the platform. any extension needs to request explicitly to get activated, or else it simply is not loaded. so you are not loading most of the IDE until its actually needed.
i too haven't heard any one talking about moving away from OSGi. and i agree, moving away from OSGi isn't practical too. but OSGi with lazy load enforced at the platform is is different from the current OSGi. lazy activation existed since long ago. The fact
that it hasn't solved Eclipse's startup performance problem in ~20 years is an evidence on the architectural improvement opportunity. in summary, bold experiments would not happen without bold thoughts, and major enhancements (such as fast bootstrap) will
not happen without major architectural shifts either.
This Message Is From an External Sender
This message came from outside your organization.
> OSGi classloading is slow due to one loader per plugin, among other reasons. JPMS is light weight and can load classes faster. can v4.x accommodate / entertain this change? even as an experiment?
I am not sure whether anyone would be willing to move away from OSGi in any capacity. If there is a possibility to change equinox to use module layers with one class loader for some of the bundles
(if there are no split packages or other things, possibly with tycho automatically creating a module descriptor for bundles where this is possible), I guess that could be worth an experiment but it would be quite a bit of effort (though probably nothing in
comparison to rewriting all bundles). Since you mentioned startup time here, you might want to take a look at this discussion I created which is about using JDK features for better startup time:
https://github.com/eclipse-platform/eclipse.platform/discussions/2060 However, I think that parts of that may work better with a somewhat stable classpath and it might be useful to get in touch with the JDK team to find a solution where Eclipse/OSGi can
tell the JDK about that in some way (e.g. where these classes come from, possibly one archive per class loader).
I don't think getting rid of OSGi is reasonable even for a v5 if that requires all plugins to change without there being a compatibility layer of some sort.
> - plugin registry loading is heavyweight and slow because all the installed plugins are loaded up front. can we do lazy activation in v4.x?
I am not sure but I thought that at least some plugins are loaded lazily? See also
https://stackoverflow.com/a/17592449/10871900. I guess what might be needed here is making sure that it is actually used where it is possible?
On 27/05/2026 18:01, GIREESH PUNATHIL wrote:
Thank you Daniel! a couple of examples:
- OSGi classloading is slow due to one loader per plugin, among other reasons. JPMS is light weight and can load classes faster. can v4.x accommodate / entertain this change? even as an experiment?
- plugin registry loading is heavyweight and slow because all the installed plugins are loaded up front. can we do lazy activation in v4.x?
there are many ways through which we could make eclipse light weight, above are just few examples.
when you say lightweight is a "matter of willingness", pls remember none of the v4.x releases were willing to do this, and we kept on losing ground. and "we could do this in v4.x itself but havent" has become indistinguishable from "we won't." a major version
on the other hand creates the forcing function and opens up opportunities to experiment.
why do you think eclipse agent project is slow moving? IMO it is due to the complexity of the code base. a light weight IDE would make AI integration seamless.
but from the fact that eclipse agent project is slow, do you allude that eclipse users do not want AI SDK? and from a lot of people don't was AI assistance in SDK, so does the majority of the user base?
This Message Is From an External Sender
This message came from outside your organization.
Hi,
> * an IDE that is lightweight and fast. why this is needed? because all the competing IDEs are light and fast. are they more attractive because they are light and fast? yes. This requires breaking
changes with a major version.
Can you elaborate what "breaking changes" would be needed for this? It is already possible to remove plugins from the default packages (e.g. Mylyn) to make parts of it faster and make some changes
to the default perspectives for a more "lightweight" experience (that is just a matter of willingness). When it comes to performance, there are probably a lot of optimizations that could be done without breaking compatibility (which haven't been done yet).
Furthermore (on the "lightweight" side), it may also be possible to refactor some things into different OSGi bundles if most users don't need a part of the functionality and keep the existing bundle with the extracted bundles as dependencies to preserve compatibility.
> * an IDE that AI assistants can easily integrate into. major LLM SDKs build first class integrations for other IDEs. AI assisted development is the new baseline and when we don't have a credible
AI story. developers switch IDEs and do not return. and we can't do modern AI integration cleanly onto a 20 year old plugin architecture. This requires breaking changes with a major version.
I think the Eclipse Agents (https://github.com/eclipse-agents/eclipse-agents/)
proposal/project already does that but there isn't sufficient interest to actually put it forward. I don't see any reason why any of this would require any breaking changes.
What exactly are the APIs you need to make breaking changes for and how "breaking" are they? How would you get most of the existing plugins working with v5 (including ones that may not be maintained)?
Could you point out what deprecated APIs (or design decisions) are actually causing the issues you mentioned? From what I can see, both of the issues are things that could be addressed without major breaking changes but I don't feel like having a v5 in sight
will make people more willing to implement these things. I think the main problem is not being bound by staying on v4 but by the development capacity for improvements. And if minor removals are needed, these are probably still possible.
That being said, there are a lot of people who don't want AI assistents at all on their systems. Building the tools they use around AI assistents would likely cause annoyances for those people.
Yours,
Daniel
On 27/05/2026 15:37, GIREESH PUNATHIL via eclipse-dev wrote:
thanks Aleks for the links. the incremental API removal policy looks good. i did some random search and found APIs that are 15+ years old living with deprecated tag. if the policy is working smoothly, why are some of those APIs still present? i guess the answer
is API removal is hard when they are fused into the core design of the IDE.
and that was just one item from the list of concerns (breaking changes accumulating over time, exponentially harder for future migrations, introduction of awkward workarounds...) i raised.
to me, what are the compelling case for v5? what are things that cannot be developed without breaking v4?
* an IDE that is lightweight and fast. why this is needed? because all the competing IDEs are light and fast. are they more attractive because they are light and fast? yes. This requires breaking changes with a major version.
* an IDE that AI assistants can easily integrate into. major LLM SDKs build first class integrations for other IDEs. AI assisted development is the new baseline and when we don't have a credible AI story. developers switch IDEs and do not return. and we can't
do modern AI integration cleanly onto a 20 year old plugin architecture. This requires breaking changes with a major version.
This Message Is From an Untrusted Sender
You have not previously corresponded with this sender.
i want to resume this discussion , but i lost the old mail thread (more than an year old), hope the title will help sync with the thread in the email discussion archive.
what do people think about moving to v5?
i strongly support this.
What would be the driving new feature/api/design for v5? If such a development is happening yet - it would better be shared with the community so more context could be put in the discussion.
At least I'm not aware of anything like that.
if backward incompatibility and ecosystem breakage is a concern, the one time manual efforts for migration is now drastically reduced with several AI tools available.
on the other hand, if we don't move, breaking changes accumulate over time, making it exponentially harder for future migrations, while code gets filled with awkward workarounds, old APIs still linger while better designs are possible....
That's totally not the case:
Are you actually looking for the "marketing" effect of v5? I ask as so far I fail to spot any technical reason for v5.
